MULTI-IONIC LITHIUM SALTS FOR USE IN SOLID POLYMER ELECTROLYTES FOR LITHIUM BATTERIES
Genre
Thesis/DissertationDate
2015Author
Chinnam, Parameswara RaoAdvisor
Wunder, Stephanie L.Committee member
Wayland, Bradford B.Zdilla, Michael J., 1978-
Li, Christopher
Department
ChemistrySubject
Chemistry, PolymerChemistry
Lithium Batteries
Multi-ionic Lithium Salts
Polymer Lithium Batteries
Solid Polymer Electrolytes
Permanent link to this record
http://hdl.handle.net/20.500.12613/2692Metadata
Show full item recordDOI
http://dx.doi.org/10.34944/dspace/2674Abstract
Commercial lithium ion batteries use liquid electrolytes because of their high ionic conductivity (>10-3 S/cm) over a broad range of temperatures, high dielectric constant, and good electrochemical stability with the electrodes (mainly the cathode cathode). The disadvantages of their use in lithium ion batteries are that they react violently with lithium metal, have special packing needs, and have low lithium ion transference numbers (tLi+ = 0.2-0.3). These limitations prevent them from being used in high energy and power applications such as in hybrid electric vehicles (HEVs), plug in electric vehicles (EVs) and energy storage on the grid. Solid polymer electrolytes (SPEs) will be good choice for replacing liquid electrolytes in lithium/lithium ion batteries because of their increased safety and ease of processability. However, SPEs suffer from RT low ionic conductivity and transference numbers. There have been many approaches to increase the ionic conductivity in solid polymer electrolytes. These have focused on decreasing the crystallinity in the most studied polymer electrolyte, polyethylene oxide (PEO), on finding methods to promote directed ion transport, and on the development of single ion conductors, where the anions are immobile and only the Li+ ions migrate (i.e. tLi+ = 1). But these attempts have not yet achieved the goal of replacing liquid electrolytes with solid polymer electrolytes in lithium ion batteries. In order to increase ionic conductivity and lithium ion transference numbers in solid polymer electrolytes, I have focused on the development of multi-ionic lithium salts. These salts have very large anions, and thus are expected to have low tanion- and high tLi+ transference numbers. In order to make the anions dissociative, structures similar to those formed for mono-ionic salts, e.g. LiBF4 and lithium imides have been synthesized. Some of the multi-ionic salts have Janus-like structures and therefore can self-assemble in polar media. Further, it is possible that these salts may not form non-conductive ion pairs and less conductive ion triplets. First, we have prepared nanocomposite electrolytes from mixtures of two polyoctahedral silsesquioxanes (POSS) nanomaterials, each with a SiO1.5 core and eight side groups. POSS-PEG8 has eight polyethylene glycol side chains that have low glass transition (Tg) and melt (Tm) temperatures and POSS-phenyl7(BF3Li)3 is a Janus-like POSS with hydrophobic phenyl groups and -Si-O-BF3Li ionic groups clustered on one side of the SiO1.5 cube. The electron-withdrawing POSS cage and BF3 groups enable easy dissociation of the Li+. In the presence of polar POSS-PEG8, the hydrophobic phenyl rings of POSS-phenyl7(BF3Li)3 aggregate and crystallize, forming a biphasic morphology, in which the phenyl rings form the structural phase and the POSS-PEG8 forms the conductive phase. The -Si-O-BF3- Li+ groups of POSS-phenyl7(BF3Li)3 are oriented towards the polar POSS-PEG8 phase and dissociate so that the Li+ cations are solvated by the POSS-PEG8. The nonvolatile nanocomposite electrolytes are viscous liquids that do not flow under their own weight. POSS-PEG8/POSS-phenyl7(BF3Li)3 at O/Li = 16/1 has a conductivity, σ = 2.5 x 10-4 S/cm at 30°C, 17 x greater than POSS-PEG8/LiBF4, and a low activation energy (Ea ~ 3-4 kJ/mol); σ = 1.6 x 10-3 S/cm at 90°C and 1.5 x 10-5 S/cm at 10°C. The lithium ion transference number was tLi+ = 0.50 ± 0.01, due to reduced mobility of the large, bulky anion and the system exhibited low interfacial resistance that stabilized after 3 days (both at 80°C). Secondly, solid polymer electrolytes have been prepared from the same salt, POSS-phenyl7(BF3Li)3 and polyethylene oxide (PEO). These exhibit high ambient temperature conductivity, 4 x 10-4 S/cm, and transference number, tLi+ = 0.6. A two-phase morphology is proposed in which the hydrophobic phenyl groups cluster and crystallize, and the three -BF3- form an anionic pocket, with the Li+ ions solvated by the PEO phase. The high ionic conductivity results from interfacial migration of Li+ ions loosely bonded to three -BF3- anions and the ether oxygens of PEO. Physical crosslinks formed between PEO/Li+ chains and the POSS clusters account for the solid structure of the amorphous PEO matrix. The solid polymer electrolyte has an electrochemical stability window of 4.6 V and excellent interfacial stability with lithium metal. In order to further enhance the ionic conductivity of solid polymer electrolytes, we have made two improvements. First, we have used so called half cube structures, T4-POSS, that contain 4 phenyl groups on one side of a Si-O- ring, and 4 ionic groups on the other side, and so are true Janus structures. They contain a 4/4 ratio of phenyl/ionic groups, unlike the previous structures that contain 7 phenyl groups/3 ionic groups. At the same O/Li ratio, the ionic conductivity of [PhOSi(OLi)]4 with POSS-PEG8 is higher than POSS-phenyl7Li3 because of more Li+ dissociation in the former case. Second, we have increased the dissociation of the lithium salts by replacing the Si-O-BF3Li groups with Si-(C3H4NLiSO2CF3)4. Both T4-POSS-(C3H4NLiSO2CF3)4 and POSS-(C3H4NLiSO2CF3)8 have been synthesized and characterized, with some preliminary conductivity data obtained.ADA compliance
For Americans with Disabilities Act (ADA) accommodation, including help with reading this content, please contact scholarshare@temple.eduCollections
Related items
Showing items related by title, author, creator and subject.
-
HIGH MODULI POLYMER GELS AND NON-AQUEOUS ELECTROLYTES WITH MUTLI-IONIC LITHIUM SALTS HAVING HIGH THERMAL STABILITY AND LITHIUM ION TRANSFERENCE NUMBERSRechargeable Li-ion batteries play pivotal role in the growth of the market for portable electronic devices like mobile phones, laptops etc. Several key drawbacks with the Lithium-ion Batteries (LIBs) put a limitation on their use in more advanced applications especially, those that require more power output and rapid charging times such as electric vehicles. One issue is the safety concerns that arise due to the growth of lithium dendrites especially at rapid charging conditions, leading to fire hazards at extreme thermal runaway scenarios. Another issue is that the current state of the art LIBs are fast approaching their maximum theoretical energy density limits. The energy density of the present-day LIBs is insufficient to deliver satisfactory performance especially when used in advanced applications such as electric vehicles which require long driving range per charge. One possible approach suggested to increase the energy density of the battery is by replacing intercalated graphite anode with lithium metal which has greater theoretical capacity. Utilizing lithium metal would open the possibility to use cathodes with conversion chemistry that would store more charge per cycle and there by resulting in the overall increase in the energy density of the battery. One of the key issues impeding the commercialization of Li-metal batteries is safety. Safety issues arises due to the uneven deposition of Li on the surface of the Li metal which would give rise to needle like protrusions (dendrites) that are capable of piercing the separator and reaching all the way to cathode. The result is an internal short circuit of the cell , which ultimately results in fire or explosion. There are many solutions that have been proposed over the years, courtesy of some extensive theoretical and experimental research to tackle the issue of dendrite growth. This thesis describes about the research work along the lines of a couple of possible approaches to prevent the dendrite growth. The first approach to be discussed is the development of solid electrolytes/separators to address this safety issue. The major drawback with most of the solid electrolytes/separators is that they have very low conductivity and lithium ion transference number (tLi+)values. Transference number of an ion by definition is the fraction of current carried by that particular ion, here in the case of LIBs is Li+, out of the total current carried by all the ions in the solution. Ideally a tLi+ of unity is aspired. The goal was to develop solid electrolytes having good room temperature conductivity and high transference numbers and at the same time having good mechanical strength. To that extent high moduli polymer gel electrolytes with high thermal stability were developed by in-situ encapsulation of ionic liquids and solvate ionic liquids into nanofibrillar methyl cellulose networks. The separators/iongels prepared possessed good room temperature conductivity and lithium ion transference numbers. The other approach described is developing non-aqueous liquid electrolytes having high ionic conductivity and high Li+ transference numbers (tLi+). Electrolytes with high tLi+ are efficient in minimizing the concentration polarization and ultimately mitigating the dendrite growth. As a part of this approach a functionalized symmetric, multi-ionic polyhedral oligomeric silsesquioxane (POSS) with dissociative lithium salt (POSS-(LiNSO2CF3)8 ) salt was dissolved in tetraglyme (G4), CH3–O–(CH2CH2O)4–CH3 in a specific O/Li ratio and the solution mixture was used as a electrolyte. Good ionic conductivities (σ10-4 S cm-1) and lithium ion transference numbers of tLi+= 0.65 are achieved in these electrolyte systems.
-
Generation of Novel Solid-State Electrolyte: Co-Crystals of Organic Solvent and Lithium (or Sodium) SaltsThe nature of the electrolyte plays an essential role in ionic transport of any lithium or sodium-ion battery. However, there are safety issues associated with these batteries, limiting their use in high power applications because of flammable liquid electrolytes, which can lead to fire and explosion. Solid-state electrolytes are inherently safer than liquid electrolytes, which are composed of ions dissolved in high-dielectric liquid organic solvents such as propylene carbonate or ethylene carbonate, and which are used in current lithium-ion batteries (LIBs). The safety and performance issues of lithium-ion batteries are respectively related to the flammability of the liquid organic solvent electrolytes that have a low flash point, and to the continuous growth of lithium metal dendrites on the anode, at which the lithium cations are reduced and deposited on the anode of LIBs during charging. The dendrite formation and growth on the anode of Lithium metal batteries (LMBs) is also observed on the anode of sodium metal batteries (NMBs). These issues, which impair the overall quality of both battery systems, can be addressed by developing novel electrode and electrolyte materials. This thesis has focused on the design, synthesis, characterization, and investigation of new lithium (or sodium) solid state electrolytes, co-crystalline electrolytes, as well as their application in lithium or sodium metal batteries. The first such material we have prepared, DMF: LiCl, has high conductivity for a pure soft solid lithium electrolyte. Additionally, in chapter 4, new sodium co-crystalline electrolytes of dimethylformamide (DMF) organic solvent and NaClO4 salt with different solvent salt ratios 2:1 ((DMF)2NaClO4) and 3:1 (DMF)3NaClO4 have been prepared. The synthesis, and physical and electrochemical properties of (DMF)3NaClO4 and (DMF)2NaClO4 are presented. The 3:1 co-crystal can be transformed into 2:1. The crystal structures of the materials reveal parallel channels of Na+ and ClO4- ions. The pressed pellet of (DMF)3NaClO4 has better ionic conductivity (10-4 S cm-1) than the pressed pellet of (DMF)2NaClO4 at RT, and over a broad temperature range (-77 ⁰C and 50 ⁰C), with a low activation barrier of 25 kJ mol-1. The SEM of (DMF)3NaClO4 reveals thin liquid interfacial contacts between crystalline grains, which promote ion mobility. The two materials have different melts that do not decompose, and upon cooling, they re-solidify as their solid structures, permitting melt casting of the electrolytes into thin films and the fabrication of cells in the liquid state, ensuring penetration of the electrolyte between the active electrode particles. Besides the high ionic conduction of the co-crystal of DMF and NaClO4, in chapter 6, we investigated the synthesis, and the physical and electrochemical properties of the co-crystal of adiponitrile (ADN) and NaClO4 salt. The crystal structure calculation of the material reveals 3:1 solvent salt ratio as (ADN)3NaClO4, a high ionic conductivity sodium solid electrolyte. The material possesses high thermal stability (up to 150 °C) and the ability to be melt-cast (Tm = 81 °C). The pressed pellet of (ADN)3NaClO4 has a high ionic conductivity of 2.2 × 10-4 S cm−1 at RT with a low activation barrier for ion conduction of 22 kJ mol-1. The high conductivity is the result of low-affinity ion-conduction channels in bulk, based on the X-ray crystal structure, and thanks to low grain boundary resistance, as well as possibly a grain-boundary percolating network due to a fluid-like nano liquid layer between the grains, observable by scanning electron microscopy and differential scanning calorimetry. When the liquid nanolayer is rinsed away or removed by excessive drying, the bulk room temperature ionic conductivity is 4 × 10-5 S cm-1 with an activation energy of 37 kJ mol-1, and the sodium ion transference number is 0.71. Scanning electron microscopy and classical molecular dynamics simulations suggest that these cocrystals form a fluid layer of ADN at the surface, which facilitates the Na+ ion migration between the grains. Density functional theory calculations are consistent with the possibility of ion conduction via a solvent−anion coordinated transition state through vacancy defects in the three symmetry equivalent ion channels along separate directions, suggesting the possibility of ionic conductivity in three dimensions. The last sodium electrolyte in this thesis (ADN)3NaPF6), in chapter 9, is investigated by co-crystallizing the organic solvent ADN with NaPF6 salt. The calculated crystal structure shows a ratio of 3:1 AND/NaPF6 salt. The stability of the material to sodium metal and its thermal and electrochemical properties show its application to sodium metal battery. The sodium ion is solvated by six -C≡N of the ADN. (ADN)3NaPF6 presents 3D linear parallel ionic channels of Na+ and PF6- ions, with distances between two successive Na+ (and PF6-) of 8.393, 8.393, and 11.527 Å, where the shortest distance between two successive Na+ in the b-crystallographic direction is 8.39 Å. The presence of 3D channels and the large gap between two Na+ ions in the complex may facilitate the migration of Na+ in the matrix. The (ADN)3NaPF6 cocrystal melts around 98 °C. The conductivity of the pressed pellets is 2 x 10-4 S/cm at RT, with an activation energy of 38.2 kJ mol-1. The CV scans and plating tests indicate that (ADN)3NaPF6 is electrochemically stable to sodium metal up to ~ 4V, meaning the crystal can be used in a sodium metal battery. This thesis also focuses on lithium co-crystalline electrolytes by co-crystallizing Lewis base organic solvents with common lithium salts used in electrolyte materials. In chapter 5, a soft solid crystal composed of isoquinoline (IQ) and LiCl was prepared based on the concept of Pearson’s hard-soft acid-base (HSAB) theory, and in addition to the crystal structure, the thermal and ionic conduction are investigated. Single-crystal X-ray diffraction best described the (IQ)3 •(LiCl)2 as consisting of molecular Li4Cl4(IQ)6 units, where the LiCl cluster is an array of edge-fused Li2Cl2 rhombs. The pressed pellet conductivity showed that ionic mobility occurs mainly through the bulk via a hopping mechanism, with a calculated activation energy of Ea = 67 kJ mol-1. The high value of the activation energy was due to Li4Cl4 clusters that were well separated by intervening IQ ligands in the crystal structure, requiring long hops for ions to migrate through the lattice. Another lithium co-crystalline material is investigated, in chapter 7, by co-crystallizing ADN, a highly thermally and electrochemically stable organic solvent with LiPF6, a thermally unstable salt. The physical and electrochemical properties of the material are experimentally and theoretically investigated. The calculated crystal structure shows a 2:1 solvent salt ratio as co-crystalline (ADN)2LiPF6. The complex forms linear parallel lithium channels, through which the Li+ ions can migrate. The Li+ ion is solvated by four -C≡N of the ADN, with no contact ion pairs with PF6-. High conductivity (σ ~ 10-4) results from weaker interactions between “hard” Li+ ions with “soft” -C≡N. Plane-wave DFT calculations show that the mechanism of ion migration is through the formation of an intermediate between two adjacent Li+ sites in the lattice and not through a hopping mechanism as is observed in inorganic ceramic electrolytes. A liquid-like layer is found at the grain boundaries that merge the grains, so that pellets are easily formed and do not require high pressure/temperature treatments to achieve high conductivities, as in the case of ceramics. The solid (ADN)2LiPF6 has a wide electrochemical stability window of 0 to 5 V. Li0/(ADN)2LiPF6/LiFePO4 half cells exhibit cycling for > 50 cycles at C/20, C/10, C/5 rates with capacities of 140 mAh/g to 100 mAh/g and efficiencies > 95%. In line with co-crystalline (ADN)2LiPF6, in chapter 8, we co-crystallized ADN organic solvent with two LiPF6 homologous salts, LiAsF6 and LiSbF6, and explore the physical and electrochemical properties of the three material. The new co-crystalline structures have the same solvent salt ratio 2:1 as (ADN)2LiPF6. The three materials have linear parallel lithium channels in their b crystallographic direction. In each complex, the Li+ ion is solvated by four -C≡N of the ADN, with no contact ion pairs to anions. Temperature-dependent ionic conductivities (σ) and lithium-ion transference numbers (tLi+) reveal that increasing the molar mass of the salt promotes lithium mobility in the matrix, meaning that there is a more significant contribution from the Li+ ion to the conductivity as the anion becomes larger (greater mass). TGA data indicates that co-crystalline (ADN)2LiPF6 thermally stabilizes the LiPF6 salt, which decomposes at low temperatures. In the case of (ADN)2LiAsF6 and (ADN)2LiSbF6, the thermal stability of the complexes is similar to that of ADN. The three compounds melt approximately at the same temperature, around 180 ºC. The electrochemical stability window of the complexes decreases as the electronegativity of the center atom (P, As, and Sb) decreases.
-
AUTLER-TOWNES SPECTROSCOPY OF THE LITHIUM DIMER MOLECULEThis thesis consists of two experimental applications of the Autler-Townes (AT) spectroscopy. In the first experiment, we have determined the electronic transition dipole moment for the 7Li2 A1Σu+ - X1Σg+ system experimentally by using a 4-level continuous wave extended Λ excitation scheme and compared our results with theoretical predictions. 7Li2 is a good test case for the accuracy of the AT splitting based technique to determine the transition dipole moment and its internuclear distance R dependence. The molecule has only 3 electrons per atom. The A1Σu+ - X1Σg+ potential energy curves were well known and thus, one could calculate accurate rovibrational wavefunctions for the simulations. In addition two different quantum mechanical models were available for the comparison: an all-electron valence bond self-consistent-field method and a pseudo-potential molecular orbital method. Our experimental results for the absolute magnitude of the transition dipole matrix elements for rovibronic transitions for different R-centroid values are in excellent agreement with ab initio theoretical calculations of the transition dipole moment. We believe that this technique will become an important method for accurate measurement of the absolute value and R-dependence of electronic transition dipole moments in molecules. The comparison with theory reinforces this view on the accuracy and universality of the AT method. The focus of the second part of this thesis is on experimentally controlling the singlet-triplet character of the 7Li2 molecule by using an external coupling laser field. We have demonstrated experimentally for the first time that the frequency domain quantum control scheme developed by T.Kirova and F. C. Spano (Physical Review A, 71, 063816, 2005) can be used to control the mixing coefficients of a weakly perturbed pair of singlet and triplet rovibrational levels. The coupling field, when tuned to resonance with the rovibronic transition involving the singlet component, causes it to AT split, leading to enhanced mixing of the pair of levels, as predicted by theory.
